One of my daily conflicts is organized in one place organized at one place. This is the same with my sister who works on machine learning research projects at the intersection of dental science. My youngest brother is a teacher, and he has more teaching material folders on his desktop as I have patience to count.
For us, it is a task to collect the sections of learning or research materials, sources, and nuggets of notes, and united of them. After trying its proper part of the organization tools and productivity shortcuts, I finally landed on Google’s notebooklm last year. Yes, it has a lot of AI. And no, it will not overwhelm you with the facts and AI hallucinations.
Unlike chatbott like Gemini or Chatgate, the notebook form can work perfectly with its content. Then it does more. A lot, really. It can prepare your random content in well-designed documents, create a mind-map, and even make a podcast of it. You can also interrupt two hosts while discussing your written ideas, such as it was a two-person news panel.
So far, the notebooklam remains exclusive to the web platform. This made it cumbersome to access through a phone. On the eve of Google I/O, the app finally landed on mobile. And even if it still has some gaps to fill it, it can already do much more than an average note taking app. More, to be fair.
Start with notebook
The mobile app is quite basic. To keep things simple, call it nakebone, or deliberate step. You start making a notebook, which lets you add the source material. This can be a PDF stored on your phone, a YouTube video, a web article, or even a copied text from your clipboard.
Once made the notebook, the app processes all sources and is ready to answer your questions. Now, these can be hyperspective, or just wide questions. For example, I uploaded about half a dozen research papers and market analysis reports, discussing the impact of tariffs on graphite supply and their direct impact on the global EV industry.
My broad requests usually involve turning all source materials into a small article for a quick observation. However, the notebooklam can also provide needle-in-the-hestac querry, as well as with appropriate quotes. For example, when I asked it about the country that would be the worst hit, it gave me an accurate answer with additional reference.
The best part? This source links to a specific section in the material (which opens as a pop-up window) so that you can verify whether AI has drawn the correct information. In my tests, knowledge extraction has mostly been at the point, as long as you are not working with artistic materials like poetry, where metaphors can sometimes throw AI and its understanding.
You can add more sources to a notebook, and AI will summarize and tweek its reactions based on fresh learning content accordingly. Finally, in the lower bar, you have a studio section for podcast, but later more on it.
Miss something, with an easy fix
Now, Notebooklm is recalling a healthy few features that are available on the web version. For example, you cannot add your own ideas to the notebook, or turn it into a source. For this, a work -raound is locally saving your note as a PDF and then importing it to the notebooklm app.
One of the most complicated characteristics of Notebooklm is making Mindmaps, but they are also missing from the mobile app. Similarly, you cannot customize or adjust the length of the podcast in the app. Finally, the options to generate study guides, briefing doors, FAQ, and timeline are also a-show.
Thankfully, you can do all this in a mobile browser. Once you make a FAQ or briefing doc, just add it as a source with a single tap, and you can access it on the mobile app. The only exception is Mind Maps, as they are saved as PNG, a file format that is not currently supported to upload on the mobile app. This trick is currently reserved for Mithun, although I hope it is expected to land on the notebook soon.
Podcast is the rear winner
One of the standout features of Notebooklm is the native podcast generation. You can simply upload all your sources URL, PDF files and notes, and AI on the ship can generate a two-person podcast for you. These podcasts have made the process of learning and modification a more immersive, especially for a person like me, who stares the text throughout the day.
I recently had a discussion about prep work for an interview with my colleagues. It has happened quite often that despite preapping in advance, I forget one or two major talk points. This time, I passed through crowd -Sau -Sour questions as an interactive podcast, and made a permanent impression as a regular list of bullet points.
But there is much more for these podcasts. You can also interrupt the hosts and ask them relevant questions about the subject being discussed. It is an underraged perk for two reasons, especially at the age of AI. First of all, you know where the audio clips are getting your content.
Second, you do not have a burden with the trust condrum of talking with AI, with confidence to spew garbage with confidence like applying glue on a pizza recipe.
See, there is no denying that the Internet as we know it has quickly become a dumping ground for AI slope. Google is partially guilty of this. Features such as AI overview and AI search mode still struggle briefly or even wrongly misunderstand the basic facts.
Similarly, YouTube and other social media platforms are rapidly bombing with a lot of Asuchi claims and lowly misleading information with AI-borne clips. The choice of Spotify and Amazon has also loosened its stance on AI content. In short, this is the burden of your facts.
The podcasts generated by the notebooklam avoid that dilemma. What you hear from the hosts is purely what you supplied them in the first place. YouTube video from colleague research paper, reliable sources, articles, or your own music (with all grammatical errors in tow).
Now, the scientific community is divided on whether listening is definitely better than looking to absorb knowledge. A linguistics specialist and teacher told me that a hybrid learning attitude is better. Since it attaches our senses more, the learning process is a little more impervious and low drab.
Of course, no one can exempt the power of creative persuasion when it comes to learning complex subjects. By that end, being able to stop the podcast host and ask them deeply questions. And when they are answered in terms of particularly supplied materials, it is presented opaquely from the web instead of an AI, you believe that the answers are reliable.
It should be on your phone
The notebooklam is what you will call the future of taking a note. This is an app that essentially converts your notes (and all the material you have collected) into an interactive format. A format where you can chat a back and forth with an answer-giving machine, which is to swallow all the knowledge you supplied.
It goes one step forward and then converts it into a podcast where even the most technical papers turn into a fun two-person audio conversation. You get the flexibility of converting all your reading content into a variety of formats, which is ready to share or individual reading. I like me as a FAQ, and with the same tap.
Right now there is only Miss app. I am not sure why this is so. Technically, the mobile browser version of the Notebook Comm is all you need, but it is still a bit of friction. Yes, you can easily create a web app shortcut and achieve those boundaries.
Google says that the notebooklm has found a lot of traction among the students, and I can definitely do it in my home. But I think your phone is ready for all that is on-device Gemini Nano Pizza, you should reach it all. This year, depending on what I saw in Google’s I/O, it could happen soon.
